topic-brainstormer
Skill448 starsChanged 5 days agoMITView on GitHub
Generate blog topic ideas: problem mining, gap analysis, expansion.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- notque/claude-code-toolkitword for word, first seen Oct 8, 2026
Other versions1 project uses a version of this.
Word for word · 1
- notque/claude-code-toolkit★ 448100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository64 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- assessmentSkill
- dataSkill
- business-opsSkill
- code-qualitySkill
- comment-qualitySkill
- condenseSkill
- joy-checkSkill
- content-engineSkill
- contentSkill
- headlinesSkill
- image-genSkill
- image-to-videoSkill
- series-plannerSkill
- video-editingSkill
- writingSkill
- domainSkill
- cli-designSkill
- frontendSkill
- frontend-slidesSkill
- jev-designSkill
- webgl-card-effectsSkill
- game-asset-generatorSkill
- game-devSkill
- game-trailerSkill
- motion-pipelineSkill
- browser-jev-automationSkill
- cve-source-checkSkill
- deploySkill
- endpoint-validatorSkill
- headless-cron-creatorSkill
- kubernetesSkill
- githubSkill
- accountsSkill
- auto-dreamSkill
- building-with-jevSkill
- codexSkill
- docs-sync-checkerSkill
- doSkill
- dSkill
- grill-jevSkill
- toolkitSkill
- workflow-helpSkill
- debuggingSkill
- plant-seedSkill
- processSkill
- pr-workflowSkill
- quickSkill
- testingSkill
- with-anti-rationalizationSkill
- workflowSkill
- worktree-agentSkill
- programmingSkill
- architecture-deepeningSkill
- codebase-analyzerSkill
- markdown-converterSkill
- news-collectionSkill
- researchSkill
- roastSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 45
Read all posts in the content directory.
- Distinctive: line 78
Tie-breaking: prefer topics that fill an existing gap, complement recent posts, use already-covered technologies, have clearer narrative …
- Distinctive: line 80
Gate: All topics scored, prioritized, presented with recommendations.
- Distinctive: line 84
When the task asks for angles on a single topic (not fresh candidates), apply these six lenses instead of the content quality filter:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Topic Brainstormer
Generate blog topic ideas that align with a content identity built around solving frustrating technical problems. Three phases: ASSESS, DECIDE, GENERATE. Every topic must pass the three-question content quality filter before presentation.
Deep References
| Signal | Load | Content |
| Filtering topics: three-question test, category examples | references/content-filter.md | Full quality filter with green/red/yellow topic categories, salvage patterns |
| Scoring: impact, vex, resolution matrix | references/priority-scoring.md | Rubrics per dimension, score ranges, calibration, worked examples |
| Mining: problem sources, gap analysis, expansion | references/topic-sources.md | Mining prompts, signal strength, gap types, expansion strategies |
Phase 1: ASSESS
Gather context about existing content and available sources.
- Read all posts in the content directory. Document post count, content clusters, technologies covered, last post date.
- Identify available sources: problem mining (recent debugging, errors, config struggles), gap analysis (cross-references leading nowhere), tech expansion (adjacent technologies).
- Extract all "see also" and cross-reference mentions. Flag any pointing to content that does not exist.
Gate: Content landscape documented, at least 2 sources identified with material.
Phase 2: DECIDE
Generate candidates and filter through the content quality test.
- Mine 5-10 raw candidates from at least 2 sources. Capture: source type, raw topic, initial vex signal.
- Apply the content quality filter -- every candidate must answer YES to all three:
- Was there genuine frustration? Real time lost, failed attempts, unclear docs, unexpected behavior.
- Is there a satisfying resolution? Clear fix, understanding gained, prevention strategy, "a-ha moment."
- Would this help others? Reproducible problem, actionable solution, relatable frustration.
- Reject failing candidates. Document each rejection: topic, failed question, reason.
Failure modes to catch:
- Tutorial-only topics: "How to Set Up X" with vex listed as "learning a new tool" is not genuine frustration. Find the specific friction. "Hugo Local Build Works But Cloudflare Deploy Fails" has real vex.
- Opinion without experience: "Why Go Is Better Than Python" is debate, not experience. Ground in measurement.
Gate: At least 3 candidates pass. Fewer than 3 -- return to Step 1 with different sources.
Phase 3: GENERATE
Score, prioritize, and present the filtered list.
Scoring: Priority = Impact(1-5) x Vex(1-5) x Resolution(1-5). Thresholds: 60-125 HIGH, 30-59 MEDIUM, 15-29 LOW, 1-14 SKIP.
Titles: Replace vague categories with failure-mode titles. Bad: "Kubernetes Networking Issues". Good: "Pod-to-Pod Traffic Works But Service Discovery Fails".
Output format: Grouped by priority tier. Each topic: title, vex, joy, content cluster fit, word estimate, score breakdown. End with recommendations: top pick, quick win, deep dive.
Tie-breaking: prefer topics that fill an existing gap, complement recent posts, use already-covered technologies, have clearer narrative structure.
Gate: All topics scored, prioritized, presented with recommendations.
Angle Lenses Mode
When the task asks for angles on a single topic (not fresh candidates), apply these six lenses instead of the content quality filter:
| Lens | Question |
| Perspective shift | Whose view changes the story? |
| Ladder of abstraction | One rung up (the trend) or down (the single case)? |
| News values | Which value carries it: conflict, proximity, novelty, impact, human interest? |
| Data angle | What does the dataset say that the narrative misses? |
| Contrarian | What if the consensus framing is wrong? |
| Timeliness peg | Why now? What event or deadline makes this urgent? |
Rules: one lens per kept angle (distinctness is structural). Each kept angle must pass the "so what?" gate -- one sentence naming a concrete reader payoff. Log refused angles with reason.
Error Handling
| Error | Cause | Solution |
| No existing posts | Content directory empty | Focus on problem mining; ask about recent debugging sessions |
| All candidates fail filter | Sources lack frustration signals | Ask probing questions: "What broke recently?" Shift sources. |
| Topic too broad | Category, not a specific problem | Break into failure modes: "[A] works but [B] fails" |
| Resolution unclear | Ongoing issue, no fix yet | Defer until resolved; or assess if "understanding the workaround" suffices |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.